Estrone Sulfate in Milk During Pregnancy

This chart shows the individual estrone sulfate levels in milk of cows at various days pregnant. The assay designates pregnancy as 150 pg/mL of estrone sulfate after 120 days pregnant. This chart shows a large group of animals that is not fitting that parameter.

Milk Progesterone

This chart demonstrates the average and ranges of milk progesterone values for pregnant and open cows who are represented in the serum progesterone chart. Animals with progesterone values less than or equal to 5ng/mL are considered open.

Serum Progesterone

This chart demonstrates the average and ranges of serum progesterone values for pregnant and open cows. Animals with progesterone values less than or equal to 2ng/mL are considered open.

Signing Off

I am now in my final week of my internship. What have I learned? Twelve weeks is a very short time to complete a research project. I have also deepened my understanding of diagnostic immunology, lab techniques, and research and development in private industry.

The above charts show the final progesterone values obtained from our sample bank. Based on the results obtained from approximately 160 cows in various stages of pregnancy, we have determined that cows with progesterone levels in serum less than or equal to 2 ng/mL should be considered open. Cows with progesterone levels in milk less than or equal to 5 ng/mL should be considered open. AntelBio will soon be offering progesterone testing on DHIA milk samples. The research crew at AntelBio will continue working on creating an in-house progesterone assay. By developing one's own assay, AntelBio will be able to provide this service at a significantly reduced cost to producers.

The estrone sulfate assay data for the same cows has been turned over temporarily to the parent company to determine if the variability in results is due to the assay, improper technique in the Antel lab, or my amateur ELISA skills. The research to develop a more sensitive assay for estrone sulfate will continue. If this assay can be refined and improved it will also be offered in the future for DHIA milk samples. This would provide a great management tool for producers.

Finally, I would like to thank Todd and Bridgette for their wealth of knowledge and guidance. I wouldlike to thamk all of the AntelBio employees for their support and being such fun people to work with.

So Many Samples........

The last few weeks have kept me quite busy in the lab. I have tested many of the serum and milk samples from Nobis Dairy for estrone sulfate using the Confirm assay from Australia. The assay that had appeared so promising has some unredeeming qualities. The intra-assay and inter-assay variability is very high. The assay is not very sensitive making it difficult to distinguish estrone sulfate levels below 150-200 pg/mL. This is why the assay was only marketed for confirming pregnancy after 120 days. What we need to determine is whether this is due to a large variability in estrone sulfate levels of cows in early pregnancy, a physiological spike in estrone sulfate at 120 days of pregnancy or if the assay is not sensitive enough. This is evident in the following charts where a sharp increase in estrone sulfate is seen at 120 day pregnancies. To help answer our question, we have started using a human estrone kit from Ann Arbor. We are able to use an estrone kit because it claims 100% cross reactivity with estrone sulfate. The kit seems to be able to pick up estrone sulfate in bovine serum and milk. We have tested various dilutions with this very sensitive kit to determine what dilution yields the most accurate results. I have finally received more Ridgeway kits and I am catching up on progesterone testing. Progesterone ELISAs may have more commercial promise than originally thought, as we have already had questions from clients inquiring whether we can offer a progesterone assay cheaper than the labs they are currently using to monitor reproductive performance in cattle herds. We will also start analyzing the combined estrone sulfate, progesterone, and herd records data.

Estrone Sulfate: Milk

This chart represents the estrone sulfate levels in milk of the same individuals represented in the serum chart. At this point, the trend has been to see higher estrone sulfate levels in serum than milk, and a greater variability between individual cows in serum levels.

Estrone Sulfate: Serum

This chart represents the data we have compiled on estrone sulfate levels in the serum of lactating Holsteins. Cows are divided into early open, open, and pregnant categories. Pregnant cows are divided into groups based on the number of days they have carried a calf. Each group currently contains between 10 and 20 animals. This number will increase as we sample more animals. The data presented here represents the average, high, and low values of estrone sulfate in picograms/mL for each group of animals. Animals whose individual data points are greatly skewing the average are slated to be retested, but are currently still represented on the data seen here. (To view chart in a larger scale click on image.)

ELISA Pro?

I have spent the last two weeks fine-tuning my lab techniques and honing my ELISA skills. The first two human progesterone assays have been abandoned for the time being. We have been working with bovine serum and milk progesterone kits from the United Kingdom. This assay has worked very well, but the assay is not very sensitive and was designed for qualitative results. We are currently verifying parallelism {similarity between serially diluted samples and standard curves} within this assay with several samples from cows in late pregnancy. This looks promising after purchasing an alternate substrate that can be read by a absorbance filter in our possesion. We have also started working with estrone sulfate kits from Australia. This is a sweet little assay which appears very promising. It was specifically designed for pregnancy testing DHIA milk samples. I will be focusing on incresing the sensitivity of the estrone sulfate assay by looking at other assays (human) and possibly building our own estrone sulfate assay by purchasing indivdual components. The Australian kit is designed to confirm pregnancy at 120 days, but our goal is to lower that number.

The above graphs show a quick interpretation of the standard curves from two different human progesterone assays including the negative control bovine samples spiked with equivalent concentrations of progesterone. The Alpha Diagnostic assay appears promising, and we are currently waiting on testing of our negative control serum samples to validate low progesterone levels were present at the time those samples were taken. The Assay Designs kit is a more sensitive assay and has shown interference when bovine samples are used and an inability of the reagents to remain stable in the lab environment. (To view graphs in more detail click on image to increase size)

Introduction to ELISA Procedure

The goal of our lab work this week has been to determine which progesterone assays will work for bovine serum and milk. The two assays we have tried thus far were both designed for human serum. We first produced standard curves with progesterone samples provided in the kit or via serial dilution to determine if the assays were functioning properly. Bovine serum and milk samples were collected from cows enrolled in the G6G Ovsynch program at Nobis Dairy on days where natural progesterone levels were at their lowest concentration. These samples will provide our negative controls. The second day we repeated the standard dilutions and compared them to bovine serum and milk samples which had been spiked with equivalent concentrations of progesterone. This allowed us to look for components in milk or serum samples that may interfere with the reagents of the human progesterone assays. The assays were also run with undiluted, unaltered serum and defatted milk samples from the negative control samples. The assays which will be used for a marketable product need to perform well with unaltered samples of milk. Any extra processing required in an assay drives up production costs and will not allow us to offer a product at an affordable price to producers.

Therefore, I have begun my crash course in ELISAs. Running ELISA's consists of accurately pippetting minute quantities of reagents into small microtiter wells. This sounds fairly simple, especially when well-quipped with sophisticated equipment. I have learned that the degree of accuracy increases with the number of ELISAs the lab tech has run. This is very evident when I try to produce equivalent results when processing duplicate rows of samples compared to the lab techs that make everything look very easy when they dispense their billionth sample.

A standard 96-well ELISA plate. In this plate, bovine serum is being run alongside standards provided for a human serum progesterone assay kit to determine if the components in bovine serum interfere with the reagents of an assay designed for human serum.

This is a portion of the ELISA lab where the experiments for this project are being executed.

Many thanks to the lab techs. Pictured here: Heather Donohue and Beth Lockwood

Bridgette and I collecting blood samples.

One of our trial participants.

Collecting milk samples in the Nobis parlor. Hopefully this was not the rambunctious, uncooperative heifer group. The employees at Nobis Dairy Farms were very helpful and patient with us impeding their normally fast-paced milking routine.

My supervisor, Todd Byrem, demonstrates his love affair with the bovine species when he decides to go for a little joyride on a holstein.

Sample Collection

My first week of my internship consisted of familiarizing myself with my new place of employment and doing some background reading on the primary pregnancy hormones we will be focusing on. This includes progesterone, estrone sulfate (an estrogen metabolite) and some of the pregnancy related proteins. I have also realized the benefit of staying awake in immunology class as I am quickly becoming aware of my lack of working knowledge on ELISA tests. More on ELISA's later.

It was decided that our sample bank would be collected from Nobis Dairy Farms in St. John's, Michigan. After contacting the herd owners a schedule was arranged to collect serum and blood samples from the 700 lactating bovines. Blood and milk samples were collected during the same day from individual animals to provide a reference serum hormone level for each milk sample. Unfortunately, morning milkings at the Nobis farmstead begin at 2:30 AM. This resulted in some "zombie-like" behavior from myself, my supervisors (Todd Byrem and Bridgette Voisinet), and the willing (?) lab techs who were dragged along. Upon returning to the lab, samples were processed and stored in a freezer. Processing involves harvesting serum from the blood samples and freezing aliquots of both whole milk and defatted milk. Two types of milk samples were saved, allowing us to compare the effects of butterfat concentration on milk hormone levels. When samples were not being collected I prepared materials for collecting and processing samples. I have also started perusing the well-kept records of Nobis Dairy Farms to determine each cow's reproductive status.